A review on microelectrode recording selection of features for machine learning in deep brain stimulation surgery for Parkinson’s disease

Wan, Kong-Wah; Maszczyk, Tomasz; See, Anna; Dauwels, Justin; King, Nicolas Kon Kam

doi:10.1016/j.clinph.2018.09.018

Cited by 46 publications

(37 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In DBS surgery for other diseases, targeting is commonly aided by microelectrode recordings (MERs), where electrode positions are refined intraoperatively according to neurophysiological ‘signatures’ of the stimulation target. For example, in Parkinson’s disease, MERs show that the subthalamic nucleus is distinguishable from surrounding structures by changes in neuronal spike firing and background noise 14. Comparatively few studies have assessed the utility of MERs in localising stimulation targets for epilepsy 15…”

Section: Introductionmentioning

confidence: 99%

Targeting the centromedian thalamic nucleus for deep brain stimulation

Warren

Dalic

Thevathasan

et al. 2020

J Neurol Neurosurg Psychiatry

View full text Add to dashboard Cite

ObjectivesDeep brain stimulation (DBS) of the centromedian thalamic nucleus (CM) is an emerging treatment for multiple brain diseases, including the drug-resistant epilepsy Lennox-Gastaut syndrome (LGS). We aimed to improve neurosurgical targeting of the CM by: (1) developing a structural MRI approach for CM visualisation, (2) identifying the CM’s neurophysiological characteristics using microelectrode recordings (MERs) and (3) mapping connectivity from CM-DBS sites using functional MRI (fMRI).Methods19 patients with LGS (mean age=28 years) underwent presurgical 3T MRI using magnetisation-prepared 2 rapid acquisition gradient-echoes (MP2RAGE) and fMRI sequences; 16 patients proceeded to bilateral CM-DBS implantation and intraoperative thalamic MERs. CM visualisation was achieved by highlighting intrathalamic borders on MP2RAGE using Sobel edge detection. Mixed-effects analysis compared two MER features (spike firing rate and background noise) between ventrolateral, CM and parafasicular nuclei. Resting-state fMRI connectivity was assessed using implanted CM-DBS electrode positions as regions of interest.ResultsThe CM appeared as a hyperintense region bordering the comparatively hypointense pulvinar, mediodorsal and parafasicular nuclei. At the group level, reduced spike firing and background noise distinguished CM from the ventrolateral nucleus; however, these trends were not found in 20%–25% of individual MER trajectories. Areas of fMRI connectivity included basal ganglia, brainstem, cerebellum, sensorimotor/premotor and limbic cortex.ConclusionsIn the largest clinical trial of DBS undertaken in patients with LGS to date, we show that accurate targeting of the CM is achievable using 3T MP2RAGE MRI. Intraoperative MERs may provide additional localising features in some cases; however, their utility is limited by interpatient variability. Therapeutic effects of CM-DBS may be mediated via connectivity with brain networks that support diverse arousal, cognitive and sensorimotor processes.

show abstract

Section: Introductionmentioning

confidence: 99%

Targeting the centromedian thalamic nucleus for deep brain stimulation

Warren

Dalic

Thevathasan

et al. 2020

J Neurol Neurosurg Psychiatry

View full text Add to dashboard Cite

show abstract

“…The topic of STN localization accuracy using electrophysiology has been studied in the literature, and several techniques have been implemented (e.g., [11]- [13]). A complete review survey has been published recently on all the studies conducted so far that used different feature extraction techniques and machine learning algorithms for localizing the STN nucleus [14]. In [14], Wan et al, have reported a complete summary of the state-of-the-art algorithms that have achieved good accuracies.…”

Section: Introductionmentioning

confidence: 99%

Intraoperative Localization of STN During DBS Surgery Using a Data-Driven Model

Khosravi

Atashzar

Gilmore

et al. 2020

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

A new approach is presented for localizing the Subthalamic Nucleus (STN) during Deep Brain Stimulation (DBS) surgery based on microelectrode recordings (MERs). DBS is an accepted treatment for individuals living with Parkinson's Disease (PD). This surgery involves implantation of a permanent electrode inside the STN to deliver electrical current. Since the STN is a very small region inside the brain, accurate placement of an electrode is a challenging task for the surgical team. Prior to placement of the permanent electrode, microelectrode recordings of brain activity are used intraoperatively to localize the STN. The placement of the electrode and the success of the therapy depend on this location. In this paper, an objective approach is implemented to help the surgical team in localizing the STN. This is achieved by processing the MER signals and extracting features during the surgery to be used in a Machine Learning (ML) algorithm for defining the neurophysiological borders of the STN. For this purpose, a new classification approach is proposed with the goal of detecting both the dorsal and the ventral borders of the STN during the surgical procedure. Results collected from 100 PD patients in this study, show that by calculating and extracting wavelet transformation features from MER signals and using a data-driven computational deep neural network model, it is possible to detect the borders of the STN with an accuracy of 92%. The proposed method can be implemented in real-time during the surgery to model the neurophysiological nonlinearity along the path of the electrode trajectory during insertion. INDEX TERMS Deep brain stimulation, deep neural network, intraoperative localization of STN, Parkinson's disease.

show abstract

“…Therefore, in order to improve the efficiency and speed up this critical intra-operative localization step, many studies have focused on the automation of STN detection using MER data. These studies extracted temporal or frequency relevant features [12][13] [14][15] [16], leaving out potential additional informative features. However, the use of handdesigned descriptive features is limited by the need for normalization and extraction from the raw data, which can be problematic across centers and MER acquisition protocols.…”

Section: Related Workmentioning

confidence: 99%

Extending convolutional neural networks for localizing the subthalamic nucleus from micro-electrode recordings in Parkinson’s disease

Martin

Peralta

Gilmore

et al. 2021

Biomedical Signal Processing and Control

View full text Add to dashboard Cite

Deep brain stimulation (DBS) is an interventional treatment for Parkinson's disease which involves the precise positioning of stimulated electrodes within deep brain structures, such as the SubThalamic Nucleus (STN). Although originally identified via imaging, additional inter-operative guidance is necessary to localize the target anatomy. Analysis of Micro-Electrode Recordings (MERs) allows for a trained neurophysiologist to infer the underlying anatomy at a particular electrode position using human audition, although it is subjective and requires a high degree of expertise. Various approaches to assist MER analysis during DBS are proposed in the literature, including deep learning methods, which rely on a static input description, that is, a pre-defined number of features or input size. In this paper, we propose two dynamic deep learning approaches adaptable to the complexity of MERs signal, by using an arbitrary long listening time (in 1s chunks), while providing feedback to the neurophysiologist as to the model's certainty. We evaluated five different deep learning based classifiers which can use arbitrary length MERs for STN segmentation. We found that a Bayesian extension using the highlevel features from SepaConvNet performed the best, increasing the balanced accuracy to 83.5%. This work represents a step forward in integrating automated analysis of MERs into the DBS surgical workflow by automatically finding and exploiting possible efficiencies in MER acquisition.

show abstract

A review on microelectrode recording selection of features for machine learning in deep brain stimulation surgery for Parkinson’s disease

Cited by 46 publications

References 45 publications

Targeting the centromedian thalamic nucleus for deep brain stimulation

Targeting the centromedian thalamic nucleus for deep brain stimulation

Intraoperative Localization of STN During DBS Surgery Using a Data-Driven Model

Extending convolutional neural networks for localizing the subthalamic nucleus from micro-electrode recordings in Parkinson’s disease

Contact Info

Product

Resources

About